Method and device for the reduction of the DC component of a signal transposed into baseband, in particular in a receiver of the direct conversion type

Abstract

A method is for reducing a DC component of an input signal transposed into baseband and being generated by a first frequency transposition stage starting from an initial signal and from a transposition signal. The method includes amplifying the transposed input signal in a first amplifier. The first amplifier receives at a DC offset compensation input, a compensation signal extracted from an output signal of a second amplifier subjected to a compensation of a offset DC voltage of the second amplifier. The method also included alternating between receiving at an input of the second amplifier, a first auxiliary signal from an auto-transposition of a transposition signal in a second frequency transposition stage and a second auxiliary signal from a transposition of the initial signal in the second frequency transposition stage with the transposition signal.

Description

FIELD OF THE INVENTION[0001]The invention relates to the frequency transposition of signals, notably, the reduction of the DC component of a signal transposed into baseband, in particular, in a receiver of the direct conversion type.BACKGROUND OF THE INVENTION[0002]A direct conversion receiver, known alternatively, as a zero intermediate frequency receiver (ZIF receiver) has a homodyne architecture and converts (transposes) the received signal, for example, a radio frequency signal, directly into baseband, i.e. around the 0 Hertz frequency. The advantages of such an architecture with the homodyne structure are the elimination of both the image processing and the intermediate frequency filters. For this reason, such a receiver is easier to integrate than a heterodyne receiver in which the conversion into baseband is carried out by way of one or more intermediate transpositions at intermediate frequencies.[0003]However, a homodyne architecture may have drawbacks, which are notably the...

AI Technical Summary

Benefits of technology

"The present invention provides a method and device for reducing the DC offset of a signal in a receiver system, particularly in WCDMA systems. The method involves using an auxiliary amplifier and a main amplifier to transpose the signal and a compensation signal, followed by a Miller integrator to continuously maintain a compensation of the DC offset voltage of the auxiliary amplifier. The device includes a main input, an auxiliary input, a main amplifier, an auxiliary frequency transposition stage, an auxiliary amplifier, and a compensation means or compensator. The technical effect of the invention is to reduce the DC offset of the signal without degrading the quality of the signal around 0 Hertz."

Problems solved by technology

However, a homodyne architecture may have drawbacks, which are notably the presence in the transposed signal of a DC offset, flicker noise, a greater sensitivity to leaks from the local oscillator generating the frequency transposition signal, and also to the mismatches between the two quadrature channels I and Q of the receiver system.

The DC offset may be a major drawback in the direct conversion architecture.

The phenomenon, in fact, comprises of the appearance of a DC offset around the 0 Hertz frequency, which then leads to degradation of the performance of the receiver.

This is because such saturation causes significant distortions of the signal.

As indicated hereinabove, the self-mixing results from leaks occurring in the transposition signal generated by the local oscillator towards the signal input of the mixer or else towards the input of the low-noise amplifier generally disposed upstream of this mixer (and which may therefore amplify this leak).

Furthermore, since the transposition signal is situated within the passband of the pre-selection filter disposed just after the antenna, it may perfectly well be envisaged that the leaks reach as far as the antenna where they are subsequently returned.

In that case, the leaks from the local oscillator can interfere with other receivers in the wireless communications system, but also be reflected on external objects.

However, such a system may be slow to tune and poses problems when establishing communications.

Furthermore, the signal may be cut off at very low frequencies, which causes a degradation of the signal (loss of data).

However, such an approach may only allow the static DC offsets to be eliminated and not the dynamic DC offsets.

Moreover, since all the components are doubled up, the power consumption of the circuit increases.

However, such an approach may have a drawback for the communications systems of the CDMA or WCDMA (Wide Band Code Division Multiple Access) type in which QPSK modulation is used.

Now, such a approach with Miller integrator in negative feedback mode performs a blind suppression of the DC signal, in other words, it does not differentiate between the offset of the useful signal and the noise-signal offset.

Consequently, this may result in the DC offset noise signal being eliminated but at the expense of the quality of the received signal leading to many decoding errors and to a poor quality of reception.

Images